An agent is a specialist Claude hands a whole job to, with its own tools and its own context.
1,669 agents across 245 plugins.
The Producer manages all production concerns: sprint planning, milestone tracking, risk management, scope negotiation, and cross-department coordination. This…
The Product Manager owns the product roadmap, user requirements, PRDs, user stories, and the bridge between business goals and technical implementation. Use…
The Prototyper builds rapid proof-of-concept implementations, throwaway spikes, and time-boxed experiments to validate ideas before committing to full…
Use this agent for read-only UX evaluation of test-runner driver artifacts (Playwright AX-tree snapshots, screenshots, console output). Applies the 4-check UX…
Practitioner review agent for /craft:research-verify. Reads verified claims and challenges them from practical experience - not sources. Catches claims that…
The human-truth layer for product decisions. Consult when diagnosing why users aren't adopting, when deciding whether to iterate or kill, when interpreting…
Use this agent when initializing craft for a project or when deep project analysis is needed. Autonomously detects project type, tech stack, patterns, and…
프로젝트 스캔 전문가. 프로젝트 구조, 프레임워크, 의존성, 디자인 토큰을 초고속 탐색. Use proactively when project profile is missing or outdated.
OWASP Top 10 기반 보안 감사 전문가. REVIEW/QA 모드 보안 스캔 담당. Use proactively for security audits.
보안 감사 전문가. SEC-01~15 스캔 + OWASP Top 10 체크. Guardian Team 병렬 실행. Fail-Only 출력.
深入分析問題描述,提取關鍵資訊並制定調查方向。 適用於各種來源的問題報告: - PM 或使用者提出的線上問題 - 前端開發者反映的 UI 或互動問題 - 後端開發者報告的 API 或資料問題 - 測試團隊發現的異常行為 - 生產環境的錯誤日誌或監控告警 使用時機: - "分析這個問題的核心資訊" -…
深入分析特定的可能原因,驗證假設,定位問題的根本原因。 基於 codebase-investigator 提供的可能原因列表,從最高可能性開始逐一深入分析,直到找到確定的 root cause。 使用時機: - "驗證這個假設是否是真正的原因" - "深入分析這段程式碼的問題" - "找出這個問題的根本原因" -…
當使用者創建或修改了技能並需要品質審查、要求「審查我的技能」、「檢查技能品質」、「改進技能描述」或想確保技能遵循最佳實踐時使用此代理。在技能創建後主動觸發。範例:
Mirage detection specialist for beast-plan. Verifies plan claims against codebase reality and external facts. Catches assumptions masquerading as facts.
TDD compliance reviewer for beast-plan. Ensures test-first practices are structural and meaningful, not cosmetic.
Evaluates test quality beyond coverage. Finds zombie tests, broken tests, coupled tests, and critical untested code. Cites file:line.
Use when: next.config.* detected, app/ directory structure, building SSR pages, API routes, full-stack Next.js. Do NOT use for: pure React/Vite (no…
Use when: composer.json present WITHOUT an artisan file. Do NOT use for: Laravel apps (composer.json + artisan → laravel-expert), frontend (framework experts).
Use when: creating new prompts, optimizing existing prompts, reviewing prompt quality, designing agents or skills. Do NOT use for: code implementation (use…
Creates, optimizes, and orchestrates media assets — images (ResGrid, AVIF/WebP), video (clips, GIFs, Veo), audio (TTS, podcast), and 3D. Coordinates Cloudflare…
Master agent coordinating all tools, MCPs, agents, and skills. Plans multi-system workflows by chaining MCPs, spawning specialized agents, and using Computer…
Drizzle schema migration agent. Generates migrations from schema diffs, validates against D1/SQLite, tests rollbacks, ensures zero-downtime compatibility.
Bug root-cause investigator that traces failures, performs git archaeology, and produces diagnosis reports
Tactical code quality reviewer that identifies refactoring opportunities within existing architecture
Performance reviewer that identifies computational bottlenecks, benchmarking gaps, and optimization opportunities. Advisory only.
Reviews Unity projects for security vulnerabilities — PlayerPrefs secrets, unencrypted saves, hardcoded API keys, insecure network calls, certificate pinning,…
Creates and debugs mobile-optimized shaders — HLSL/ShaderLab, ShaderGraph custom nodes, URP shader structure, SRP Batcher compatibility, half-precision…
Writes EditMode and PlayMode tests, executes them via MCP run_tests, reports results. Knows Unity testing framework, NUnit attributes, and frame-based testing…
Controls autonomous execution mode with stop hooks and persistence
Reviews C# backend code for quality and security
Reviews Go backend code for quality and security
Security and compliance expert for vulnerability detection
Diátaxis splits documentation into four modes along two axes: **practical vs theoretical** (doing vs knowing) and **study vs work** (acquiring skill vs…
Copy this block per finding. Keep it to one anchor, one exploit sentence, one fix. Delete the guidance comments before shipping.
Doc type: **how-to guide**. Goal-first, assumes you can already run the service locally. For the concepts behind rate limiting, see the explanation doc; for…
HTML/CSS прототип для проверки user stories ДО написания production-кода. Кликабельный, главная функция работает, без backend (mock data). Цель — пользователь…
Анализ зависимостей между секциями детальной архитектуры и пересортировка в правильный порядок реализации. DAG-анализ, нет циклов, явные dependencies.
Выбор стека под bottleneck архитектуры. Учитывает 2026 tech-updates (opendataloader-pdf, VoxCPM2, chrome-devtools-mcp, llmfit) и стек по умолчанию (Next.js +…
The `high` rung of `cad-verifier` (goal-backward phase verification); `bin/route.mjs` picks it, not the user.
The `max` rung of `cad-verifier`; `bin/route.mjs` picks it, not the user.
© 2026 Flowy · Free and open source
Built for Claude Code · Not affiliated with Anthropic